MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 364: 157–167, 2008
doi: 10.3354/meps07480
Published July 29
Spawning, larval abundance and growth rate of
Sardinops sagax off southwestern Australia:
influence of an anomalous eastern boundary current
B. A. Muhling1, L. E. Beckley1,*, D. J. Gaughan2, C. M. Jones1, A. G. Miskiewicz3,
S. A. Hesp4
1
School of Environmental Science, and 4School of Biological Sciences, Murdoch University, 90 South St, Murdoch 6150,
Western Australia, Australia
2
Department of Fisheries, Western Australia, PO Box 20, North Beach 6920, Western Australia, Australia
Environment and Health Division, Wollongong City Council, 41 Burelli Street, Wollongong 2500, New South Wales, Australia
3
ABSTRACT: The temporal and spatial distributions of sardine Sardinops sagax eggs and larvae off
the oligotrophic southwestern coast of Australia were examined and related to gonadosomatic index,
daily growth rates of larvae and regional biological oceanography. Seasonal environmental cycles
were established from remotely sensed sea surface temperature and chlorophyll concentration, wind
and sea surface height data. Sardine egg and larval distributions were determined from regular
transect surveys and annual grid surveys. Sardine eggs and larvae were common across the
continental shelf throughout the year between Two Rocks and Cape Naturaliste (~32 to 34° S), and
gonadosomatic index data suggested a distinct winter peak in spawning activity. Surface chlorophyll
concentrations were highest during winter, coincident with the seasonal peak in the southward flow
of the Leeuwin Current along the continental shelf break. Retention conditions on the mid-outer shelf
for pelagic eggs and larvae were therefore poor during this time. Egg and larval concentrations were
lower than expected in winter and higher in summer when retention conditions were more
favourable. Larval sardine growth rates were unexpectedly high, averaging 0.82 mm d–1. Fisheries
for clupeiod species off southwestern Australia are insignificant compared to other eastern boundary
current systems. Our data suggest that this may be due to a combination of low primary productivity
caused by suppression of large-scale upwelling by the Leeuwin Current and the modest seasonal
maximum in primary productivity occurring during the time least favourable for pelagic larval
retention.
KEY WORDS: Biological oceanography · Sardinops sagax · Leeuwin Current · Fish larvae · Fish eggs
Resale or republication not permitted without written consent of the publisher
Small pelagic clupeiform fishes such as sardine
Sardinops sagax and anchovies Engraulis spp. are
ubiquitous in cool to warm temperate coastal oceans
worldwide, and are particularly abundant in eastern
boundary current systems. They are an important component of many marine food webs, and they support
substantial fisheries (Beckley & van der Lingen 1999).
However, recruitment is often highly variable from
year to year, resulting in large fluctuations in stock
sizes and catches (Beckley & van der Lingen 1999,
Schwartzlose et al. 1999, Smith & Moser 2003).
The sardine is generally found in upwelling areas,
with the highest catches recorded off California, Peru
and southern Africa (Beckley & van der Lingen 1999).
Much of the variability in adult population sizes and
spawning biomass of this species has been linked
to climatic and environmental variables (Beckley &
van der Lingen 1999, Schwartzlose et al. 1999). For
*Corresponding author. Email: [email protected]
© Inter-Research 2008 · www.int-res.com
INTRODUCTION
158
Mar Ecol Prog Ser 364: 157–167, 2008
Sardines occur in the eastern Indian Ocean from
inshore to outer continental shelf waters south of Shark
Bay, Western Australia (Hutchins & Thompson 2001)
(Fig. 1). However, stock sizes and commercial catches
of this species are insignificant by world standards,
with the annual catch yet to reach 20 000 t (Caputi et
al. 1996, Gaughan et al. 2001a). This is largely attributable to the lower productivity of southwestern Australian coastal waters, a product of the downwelling,
poleward-flowing Leeuwin Current (Feng et al. 2003,
Hanson et al. 2005). Leeuwin Current flow results in
southward penetration of warm, low salinity waters,
and the current is strongly seasonal, with maximum
flows in autumn and winter (Pearce et al. 2006). In contrast to the wind-induced upwelling, and consequent
increases in nutrients and chlorophyll biomass that are
characteristic of other eastern boundary current systems (Thomas et al. 2004), there is no large-scale
upwelling associated with the Leeuwin Current.
Chlorophyll biomass off southwestern Australia is,
instead, at a maximum in winter when equatorward
wind stress is at a minimum (Hanson et al. 2005,
Pearce et al. 2006, Fearns et al. 2007). This modest seasonal chlorophyll maximum likely results from winter
deepening of the mixed layer, nutrient re-suspension,
terrestrial inputs near shore (Lourey et al. 2006) and
possibly some southward entrainment of nutrients and
chlorophyll (Hanson et al. 2005).
Larval fishes feeding in southwestern Australian
neritic waters are therefore expected to grow at slower
rates than those in more productive systems in other
eastern boundary currents (Gaughan et al. 2001b), with
implications for larval fish starvation, and mortality
through predation (Bailey & Houde 1989). However, the
relationships between larval sardines and their physical
and biological environment have not been well studied
example, the stock size of sardine in the California
Current region decreased substantially from the 1940s
to the 1970s, followed by a rapid recovery in the 1980s.
Although this was partially related to fishing pressure,
a major environmental change in the northeastern
Pacific in the 1970s (from a ‘cool-ocean’ to a ‘warmocean’ oceanographic regime) resulted in marked
changes in abundance of a number of fish species,
including sardine (Smith & Moser 2003). The El Niño
phenomenon has also been implicated in changes in
sardine stock sizes in the Gulf of California (SanchezVelasco et al. 2004) and off Mexico (Funes-Rodriguez
et al. 2001). In the Benguela Current ecosystem off
southern Africa, sardine recruitment has been linked
to variability in sea surface temperature (which
approximates upwelling events), larval fish retention,
and fishing pressure (Cole 1999).
Variability in recruitment to adult sardine populations has been attributed to varying rates of survival in
early life history stages (Smith & Moser 2003). Different environmental conditions result in different retention, transport and feeding conditions for pelagic fish
larvae, with subsequent effects on recruitment (Lasker
& Smith 1976, Smith & Moser 2003). Sardines typically
spawn in continental shelf waters, and have protracted
spawning seasons (Beckley & van der Lingen 1999,
Smith & Moser 2003). The timing of spawning may differ between geographical regions (Ward et al. 2003)
and be influenced by local environmental conditions,
with maximum spawning activity occurring when
deleterious offshore transport is at a minimum and larval fish food concentrations are favourable (Somarakis
et al. 2006). Concentrations of sardine eggs and larvae
at the same location and season may therefore vary
considerably among years (Fletcher et al. 1994,
Gaughan et al. 2001a, Smith & Moser 2003).
100
200
(b)
1000
2000
m
(a)
(c)
Winter
B A Two Rocks
E D C
Hillarys
20°S
Shark Bay
Perth
32°S
Fremantle
Rottnest Island
INDIAN
OCEAN
25°S
33°S
Leeuwin
Current
(d)
Summer
Cape Naturaliste
Western
Australia
30°S
Geographe Bay
Sardine
distribution
Cape Leeuwin
114°E
34°S
Leeuwin Current
116°E
Capes Current
Fig. 1. Study area off southwestern Australia showing
(a) the overall path of the
Leeuwin Current, (b) locations of the Two Rocks
transect with stations, the
Hillarys
transect
and
places mentioned in the
text, (c) the winter and (d)
the summer variation in
location of the Leeuwin
Current. Known distribution of adult sardines is also
indicated
159
Muhling et al.: Sardine spawning, larval abundance and growth
off Western Australia, with the exception of some work
completed along the south coast, east of Albany (approximately 34 to 35° S, and 118 to 122° E) (Fletcher & Tregonning 1992, Gaughan et al. 2001a,b).
Our aim was to collate a variety of existing data sets to
determine the sardine spawning season, and temporal
and spatial distributions of sardine eggs and larvae off
the southwestern coast of Australia, and relate these to
regional biophysical oceanography. We hypothesised
that maximum abundances of sardine larvae would
occur after periods of high spawning activity (as shown
by gonadosomatic index and egg concentration data),
and periods when both feeding and retention conditions
were favourable.
MATERIALS AND METHODS
Study area. This study synthesised a number of data
sets (Table 1) collected on a range of temporal and
spatial scales off southwestern Australia between Two
Rocks (31° 35’ S) and Cape Leeuwin (34° 22’ S) between
1997 and 2004 (Fig. 1).
Environmental variables. Mean monthly ocean
colour (surface chlorophyll) data were acquired for the
study area using the GES-DISC Interactive Online
Visualization ANd aNalysis Infrastructure (Giovanni).
Although a deep chlorophyll maximum layer may be
present in our study area in summer, remotely sensed
data were considered to adequately illustrate the
regional, seasonal cycle of chlorophyll throughout the
water column. Integrated chlorophyll through the
water column is still much lower in summer than in
winter, and the seasonal cycle for depth-integrated
chlorophyll is similar to that for surface chlorophyll
data (Koslow et al. 2006, Fearns et al. 2007).
The Two Rocks transect was 84 km long, and
comprised 5 sampling stations, representing coastal
(A: 18 m depth, 31° 32.2’ S, 115° 33.6’ E), inner shelf (B:
40 m, 31° 37.1’ S, 115° 21.9’ E), outer shelf (C: 100 m,
31° 40.8’ S, 115° 13.3’ E), shelf break (D: 300 m,
31° 45.9’ S, 115° 01.2’ E), and offshore environments (E:
1000 m 31° 51.7’ S, 114° 47.6’ E). Sea surface temperatures along the Two Rocks transect were extracted
from brightness temperatures in AVHRR bands 4 and 5
from satellite images obtained by the Western Aus-
Table 1. Temporal and spatial scales of data sets analysed. DFWA: Department of Fisheries, Western Australia. GES-DISC:
Goddard Earth Sciences Data and Information Services Centre
Variable
Data source
Study area
Time period
Environmental
Ocean colour
Giovanni (GES-DISC)
Jan 1998–Dec 2004
Sea surface temperature
(a) Giovanni (GES-DISC)
31° 35’ S–34° 14’ S,
114° 30’ E–116°00’ E
(a) 31° 00’ S-34° 00’ S,
114° 00’ E–116° 00’ E
(b) Two Rocks transect:
31° 32’ S–31° 52’ S,
114° 48’ E–115° 34’ E
Fremantle:
32° 04’ S, 115° 44’ E
Rottnest Island:
32° 00’ S, 115° 30’ E
DFWA survey: Fremantle
sardine fishery
31° 00’ S–33° 00’ S
~115° 30’ E–115° 40’ E
Jan 2000–Dec 2005
DFWA grid
31° 32’ S–34° 22’ S,
114° 30’ E–115° 40’ E
Aug 2002, Jul 2004
Hillarys transect
31° 50’ S,
115° 19’ E–115° 44’ E
Jan 1997–Dec 1998
DFWA grid
31° 32’ S–34° 22’ S,
114° 30’ E–115° 40’ E
Aug 2002, Jul 2004
Hillarys transect
31° 50’ S,
115° 19’ E–115° 44’ E
31° 32’ S–31° 52’ S,
114° 48’ E–115° 34’ E
31° 32’ S–31° 52’ S,
114° 48’ E–115° 34’ E
Jan 1997–Dec 1998
(b) AVHRR bands 4–5
Fremantle Mean Sea Level
National Tidal Centre
Wind
Australian Bureau of Meteorology
Sardine spawning
Gonadosomatic index (GSI)
Sardine eggs
Sardine larvae
Sardine larvae
Two Rocks transect
Sardine larval growth
Two Rocks transect
Jan 2002–Dec 2004
Jan 1998–Dec 2004
Jan 2000–Dec 2004
Aug 2002–Dec 2004
Aug 2002–Dec 2004
160
Mar Ecol Prog Ser 364: 157–167, 2008
tralian Satellite Technology and Applications Consortium (WASTAC) (see Pearce et al. 2006). Monthly
mean temperatures for the broader study area were
obtained from Giovanni (Table 1).
Fremantle mean sea level (FMSL) data provided a
proxy for the strength of the Leeuwin Current (Feng et
al. 2003), and data were obtained from 1998 to 2004.
Wind data were obtained from an Automatic Weather
Station (AWS) operated by the Australian Bureau of
Meteorology at Rottnest Island (Fig. 1).
Sardine spawning indices. The gonadosomatic
index (GSI) (ratio of ovary weight to body weight,
DeVlamming et al. 1982) was used to calculate the sardine spawning period. Data were collected from 2730
adult females caught in the Fremantle sardine fishery
(2000 to 2005) (Table 1), which operates in inner shelf
waters off Perth (Fig. 1). As samples were obtained
from the commercial fishery, some months had more
data than others. All months had between 25 and
600 samples, except January (n = 5).
Sardine eggs (including Day 1 and Day 2 eggs) were
sampled using a grid of winter (July or August) plankton
sampling stations as part of daily egg production method
surveys completed by the Department of Fisheries,
Western Australia (DFWA) (D. Gaughan unpubl. data).
Stations were located between Two Rocks and Cape
Leeuwin from inshore waters to the shelf break (300 to
500 m deep). Eggs were collected from vertical bongo
net tows (net 0.26 m diameter, 300 µm mesh). Concentrations of eggs and larvae from all plankton samples incorporated into this study, were expressed as no. m– 3 of seawater sampled using volumes calculated from General
Oceanics flowmeters fitted to each net. Sardine egg concentrations were overlain on the monthly mean sea
surface temperature plots of the study region (data
obtained from Aqua-MODIS, NASA).
Sardine eggs < 24 h old were also counted in monthly
plankton samples (1996 to 1998) (Table 1) along the
Hillarys transect (see Pearce et al. 2006) (Fig. 1). Vertical tows were taken every 5 km with a bongo net
(0.26m diameter, 300 µm mesh) to 70 m depth, or to
within 3 m of the bottom in shallower water. Because of
rough sea conditions, some stations were omitted in
March and August 1998, and no samples were taken in
February 1997.
Larval fish. Yolk-sac, pre-flexion and post-flexion sardine larvae were removed from all vertical bongo net
tows (see above) and from oblique plankton samples
taken along the Two Rocks transect between August
2002 and December 2004 (Table 1). All stations (Stns
A–E) were sampled on a quarterly basis and, in addition,
the 3 inshore stations (Stns A–C) were sampled monthly,
when possible (see Muhling et al. 2008). Plankton
samples on the Two Rocks transect were taken with daytime oblique bongo net tows to 150 m depth, or to just
above the bottom in shallower water (Muhling et al.
2008). Nets were fitted with 100 and 355 µm mesh
(mouth area 0.196 m2, diameter 0.6 m) and were towed
at about 2 knots. Only larvae from the 355 µm mesh were
used. Plankton samples from the DFWA grid and
Hillarys transect were preserved in borax-buffered 5%
formalin, while samples from the Two Rocks transect
were split, with one subsample preserved in 10%
buffered formalin and the other in 100% ethanol.
Data on the distribution and abundance of sardine
eggs and larvae were thus available from 2 snapshots
over a broad sampling grid (DFWA samples), and from
more regular samples on a more restricted spatial scale
(Hillarys and Two Rocks transect samples).
Length and growth rate of sardine larvae. The standard notochordal lengths (SL) of sardine larvae collected at each station on each cruise along the Two
Rocks transect were measured to 0.1 mm accuracy using an eyepiece micrometer. Where > 50 larvae were
collected in any tow, a sub-sample of 50 randomly selected specimens was measured. Neuston net samples
taken on quarterly cruises along the Two Rocks transect (1 m2 square net, 1 mm mesh) provided larger sardine larvae (> 8 mm length) for growth analyses only.
Sixty-eight sardine larvae from 9 bongo net samples
(ethanol-preserved subsamples) and 2 neuston net
samples from the Two Rocks transect were aged using
daily growth rings in sagittal otoliths. The SL of each
larva used for otolith analysis was measured to the
nearest 0.05 mm using Leica IM1000 image software.
The sagittal otoliths were removed from each larva,
cleaned, air dried and mounted in cosmetic nail polish
on microscope slides. The best condition otolith from
each larva was photographed at varying foci using a
JVC TK-C1381 camera with transmitted light under an
Olympus SZX12 compound microscope. Each consecutive pair of adjacent light and dark growth zones was
assumed to correspond to daily growth zones (validated for this species by Hayashi et al. 1989). The daily
growth zones for all larvae were counted on 2 separate
occasions and without prior knowledge of standard
lengths. Where the 2 counts disagreed, a third count
was made. If all 3 counts differed, the age information
for the fish was not included in age analyses. As deposition of the initial increment for sardine larvae occurs
2.5 to 3.0 d after hatching (Hayashi et al. 1989), 2 d
were added to all counts so that post-hatch age could
be estimated.
The relationship between SL and larval age (d) was
examined and described using the Laird-Gompertz
growth curve:
αt
Lt = L0e(g0/α)(1 – e )
where Lt is the length at age t, L0 is the length at hatching (t = 0), g0 is the specific growth rate at hatching and
α is the rate of exponential decay of the specific growth
161
Muhling et al.: Sardine spawning, larval abundance and growth
05
c-
De
04
De
c-
03
c-
De
02
c-
De
01
c-
De
00
c-
De
Stn A SST
Stn C, D & E SST range
Surface chlorophyll
24
0.6
0.4
22
20
0.2
18
0
r-0
4
Ju
l-0
4
O
ct
-0
4
Ap
Ju
l-0
3
O
ct
-0
3
Ja
n04
r-0
2
Ju
l-0
2
O
ct
-0
2
Ja
n03
Ap
r-0
3
Ap
n-
02
16
Date
Fig. 3. Remotely sensed sea surface temperature (SST) and surface chlorophyll concentration (2002 to 2004) across the Two Rocks transect. Monthly
SST at Stn A, and the range of temperatures across Stns C, D and E, are
shown. See Fig. 1b for station locations
Monthly mean chlorophyll a (mg m–2)
Sea surface temperature (°C)
26
Ja
The seasonal maxima of surface chl a
and Fremantle mean sea level (FMSL, as a
proxy for Leeuwin Current strength) were
evident during late austral autumn and
winter (May to August) (Fig. 2). Strong
Leeuwin Current flow was associated with
higher surface chlorophyll, although there
was a 1 to 2 mo lag between the 2 variables
in some years, with FMSL peaking before
chlorophyll. In contrast, southerly wind
stress was highest during summer
(November to March). Southerly wind
stress and surface chlorophyll concentration were inversely correlated (R2 = 0.70)
(Fig. 2). Maximum SST across the Two
Rocks transect (2002 to 2004) occurred in
late summer (February to March), and
99
The mean monthly GSI for adult sardines sampled
from the Fremantle fishery (2000 to 2005) peaked in win-
RESULTS
Seasonal cycles of oceanographic and
biological variables
De
c-
98
De
c-
De
c-
97
FMSL (cm)
Chlorophyll a (mg m–2) /
southerly wind stress (N m–2)
120
0.7
rate. The parameters were estimated by
FMSL
maximising the log-likelihood in Excel
Surface chlorophyll
0.6
software. The Laird-Gompertz growth
Southerly wind stress
100
curve provides a significantly better fit to
0.5
the lengths at age of sardine larvae than a
80
linear curve (likelihood ratio test, p < 0.01,
0.4
Cerrato 1990). Our use of this growth
model also enabled direct comparison with
0.3
60
the results of the sardine growth study by
Gaughan et al. (2001b). The likelihood
0.2
40
ratio test was also used to determine
0.1
whether Laird-Gompertz growth curves
fitted separately to length-at-age data for
20
0
sardine larvae collected during the warmer
austral summer (November to April) dif–0.1
0
fered significantly from those for larval sardines collected over the cooler winter
period (May to October).
Date
The average daily growth rates (mm d–1)
Fig. 2. Fremantle mean sea level (FMSL) and surface chlorophyll
of sardine larvae between 3 and 18 d old
concentration (1998 to 2005)
from the lower west coast were compared
with estimates for these ages of fish (same species) in
minimum values were in late winter (August to Octosouthern Australia (118 to 137° E) using data in
ber) (Fig. 3). The surface chlorophyll cycle lagged the
Gaughan et al. (2001b). The predicted lengths at each
maximum SST signal by 3 to 5 mo. Throughout the
age were then calculated using Laird-Gompertz equastudy area, high SST and strong southerly winds in
summer were followed by increased Leeuwin Current
tions for the 2 regions. The average daily growth rates
flow and surface chlorophyll concentration through
were determined by averaging the differences in preautumn and winter.
dicted lengths between each of the successive ages.
The associated 95% confidence intervals for each of
the 2 mean values were determined using a likelihood
Sardine spawning cycles
profile approach (Hilborn & Mangel 1997).
Mar Ecol Prog Ser 364: 157–167, 2008
Mean egg concentration (no m–3)
162
n=
= 2730
2730
10
8
6
4
2
0
Jan Feb Mar Apr May Jun
Jul Aug Sep Oct Nov Dec
Month
Fig. 4. Mean monthly gonadosomatic index of adult sardines
caught by the Fremantle fishery in the Perth region (2000 to
2005). Means + SE
ter, especially in June, and was at a minimum in October
and November (Fig. 4). Sardine egg concentrations
across the Hillarys transect, however, were highest during summer (December to January) and winter (May to
September), with lower concentrations in spring (Fig. 5).
Variability in sardine egg concentrations was high between the 2 years of sampling (1997 and 1998).
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Fig. 5. Mean monthly concentrations of Day 1 sardine eggs
[ln(x+1) transformed] across the Hillarys transect (1997–1998).
Means + SE for all 6 stations (see Fig. 1b) between 15 and
40 km from shore. Some stations were omitted in March and
August 1998, and no samples were taken in February 1997.
See Fig. 1b for location of Hillarys transect
Sardine eggs were widespread between Two Rocks
and Cape Naturaliste in the winters of 2002 and 2004
(Fig. 6). During 2004, egg and larval concentrations
were highest and eggs were found at more stations
throughout the study area. Where sampling extended
22.0 Sea surface
August 2002
July 2004
20.0
temperature
(°C)
18.0
16.0
14.0
Sardine larvae
(no. m–3)
0–1
1–2
2–5
5 – 10
Sardine eggs
(no. m–3)
0 – 10
10 – 20
20 – 50
50 – 100
Sardine egg
conc. (no. m–3)
Gonadosomatic Index (%)
12
90
2002
60
30
0
16
17
18
19
20
Temperature (°C)
21
90
60
30
0
16
22
2004
17
18
19
20
Temperature (°C)
21
22
Fig. 6. Sardine egg and larval concentrations between Two Rocks and Cape Naturaliste (winter 2002 and 2004). Satellite-derived
sea surface temperatures shown
163
Muhling et al.: Sardine spawning, larval abundance and growth
1.6
Stn A
1.2
south of Cape Naturaliste, both eggs
and larvae were rare.
The majority of sardine eggs and
larvae were recorded within water of
Leeuwin Current origin (Fig. 6).
Warm water derived from the Leeuwin Current was located closer to the
coast around Geographe Bay (north
of Cape Naturaliste) in 2004 than in
2002, and this corresponded to a
more nearshore distribution of sardine eggs (Fig. 6). Comparison of the
extracted mean monthly sea surface
temperature and sardine egg concentrations for each sampled station
revealed that sardine eggs were
found in sea surface temperatures of
17 to 20.5°C. Highest egg concentrations were found at stations where
the sea surface temperature was 18 to
20°C, i.e. within water of Leeuwin
Current origin (Fig. 6).
Sardine larvae were common yearround across the Two Rocks transect
(between August 2002 and December
2004, Fig. 7). Highest larval concentrations (0.5 to 1.2 m– 3) generally
occurred on the continental shelf (Stns
B and C). Larval concentrations were
often highly variable between years
(e.g. see August and October).
2002
2003
0.8
2004
0.4
0
1.6
Stn B
Larval fish concentration (no. m–3)
1.2
0.8
0.4
0
1.6
Stn C
1.2
0.8
0.4
0
1.6
Stn D
1.2
0.8
0.4
0
1.6
Stn E
1.2
0.8
0.4
0
Jan
Feb Mar
Apr
May
Jun
Jul
Aug
Sep
Oct Nov Dec
Month
Fig. 7. Concentration (means + SE) of sardine larvae at Stns A to E on the Two
Rocks transect (August 2002 to December 2004). See Fig. 1b for station locations
80
Autumn
Summer
60
n = 259
n = 137
Winter
n = 226
Spring
n = 147
Frequency
40
20
0
50
40
30
20
>14
13 to 14
11 to 12
12 to 13
9 to 10
10 to 11
8 to 9
7 to 8
5 to 6
6 to 7
3 to 4
4 to 5
<2
2 to 3
>14
12 to 13
13 to 14
11 to 12
10 to 11
8 to 9
9 to 10
7 to 8
5 to 6
6 to 7
4 to 5
3 to 4
<2
0
2 to 3
10
Length (mm)
Fig. 8. Length frequency distributions of sardine larvae at Stns B (40 m) and C (100 m) on the Two Rocks transect (August 2002
to December 2004) by season. See Fig. 1b for station locations
164
20
Standard length (mm)
18
16
14
12
10
8
6
4
2
0
Mar Ecol Prog Ser 364: 157–167, 2008
south and west coasts of Western Australia
from June to August and from December to
February (Gaughan et al. 1990, Fletcher et
al. 1994). This is in contrast to the pattern
off South Australia, where the main sardine spawning season in summer and autumn is associated with coastal upwelling,
increased primary production and higher
zooplankton biomass (Ward et al. 2006).
Favourable feeding and growth conditions
for larvae therefore occur in conjunction
with potentially favourable larval retention
0
2
4
6
8
10
12
14
16
18
20
conditions. GSI data suggest that spawnAge (d)
ing off southwestern Australia peaks, inFig. 9. Observed data and Laird-Gompertz growth curves fitted to lengths
stead, at a time of strong southward Leeuat age of larval sardines (3 to 18 d post hatch) from the west coast of Westwin Current flow (June to August) with no
ern Australia (Two Rocks transect, Fig. 1b) during winter and summer.
upwelling. Primary productivity in the
The Laird-Gompertz growth curve for sardine larvae of the same
age range from the south coast of Western Australia is also included
study region is greatest during this time,
(Gaughan et al. 2001b)
coincident with enhanced vertical mixing.
However, given the spawning locations of
sardine (suggested by the winter egg and larval
Larval sardine lengths and growth rates
surveys) between Two Rocks and Cape Naturaliste
and the strong Leeuwin Current flow along the shelf in
Most sardine larvae caught at Stns B and C were 2
June and July (as indicated by FMSL data), it is probato 8 mm SL (Fig. 8), and small sardine larvae were
ble that eggs and larvae resulting from mid-winter
recorded throughout all seasons.
spawning events are subjected to strong southward
The Laird-Gompertz growth curves fitted to the
advection (Caputi et al. 1996). While onshore Ekman
lengths at age of larval sardines collected over the
transport associated with the passage of cold front syssummer and winter periods, i.e. November to April (n =
tems in winter may slightly improve retention of eggs
35) and May to October (n = 33), respectively, were sigand larvae and result in some interannual recruitment
nificantly different (p < 0.05). The average daily growth
variability, the monthly scale of sampling in this study
rate of larval sardines was slightly higher (0.92 mm d–1)
made the influence of such events difficult to discern.
in summer than in winter (0.78 mm d–1).
Overall concentrations of sardine larvae in the preThe likelihood ratio test indicated that the Lairdsent study were not particularly low by regional and
Gompertz growth curve for all larval sardine lengths
global standards. Across the Two Rocks transect, mean
at age estimated in this study (average daily growth
larval sardine concentrations up to 1.2 m– 3 were found,
rate = 0.82 mm d–1) was significantly different (p <
0.01) from that recorded by Gaughan et al. (2001b)
while at some southern stations on the winter sampling
for sardine larvae of comparable ages from the south
grid in 2004, larval concentrations > 2 m– 3 were frecoast of Western Australia (average daily growth
quently found. Concentrations of larvae on the Two
Rocks transect were comparable to or higher than
rate = 0.55 mm d–1) (Fig. 9). Estimates of the LairdGompertz growth parameters, L0, g0 and α for the
those found near Albany on the south coast of Western
pooled length-at-age data for the summer and winter
Australia (Fletcher & Tregonning 1992, Fletcher 1999),
periods on the west coast of Western Australia were
those measured by Beckley & Hewitson (1994) off
3.35 mm, 0.25 d–1 and 0.129 d–1, respectively; for the
southeastern Africa, and those on the open shelf off
south coast of Western Australia, these parameters
South Australia (Ward et al. 2006). Concentrations of
were 1.91 mm, 0.25 d–1 and 0.10 d–1, respectively.
larvae found at winter sampling grid stations in this
study were comparable to those in the more protected
waters of Spencer Gulf, around Kangaroo Island (Ward
DISCUSSION
et al. 2006), to concentrations in the Benguela Current
ecosystem (Huggett et al. 1998, Stevenik et al. 2001),
Spatial and temporal patterns of sardine spawning
and to concentrations of Sardina pilchardus in the
northwestern Mediterranean (Olivar et al. 2001).
Sardinops sagax spawns in shelf waters at different
However, when compared to the sardine GSI data
times in different geographical locations (Ward et al.
(from the Fremantle fishery), egg and larval concentra2003). Eggs and larvae are most abundant along the
tions (from the Hillarys and Two Rocks transects) were
Summer west coast data
Winter west coast data
Summer west coast fitted curve
Winter west coast fitted curve
Gaughan et al. (2001b)
Muhling et al.: Sardine spawning, larval abundance and growth
higher than expected in summer (given the comparatively low spawning activity), and lower than expected in
winter. Spawning activity during summer may not be as
high as in winter, but it is likely that eggs and larvae are
better retained over the shelf in summer due to the
presence of the Capes Current, a wind-induced countercurrent that flows sporadically northward over the midshelf (Gersbach et al. 1999, Pearce & Pattiaratchi 1999).
Some offshore Ekman transport may be present in summer, especially at the surface (Muhling & Beckley 2007),
but we postulate that retention conditions during
summer would still be more advantageous than during
winter. It is therefore possible that a comparatively small
proportion of the sardine spawning period could
contribute the majority of the year’s recruits to the adult
population on the lower west coast of Western Australia.
Implications for fisheries
Low commercial catches of clupeiform fishes off
southwestern Australia may be partially explained by
the oligotrophic status of the waters (Hanson et al.
2005) with consequently lowered secondary production (Koslow et al. 2006), and, therefore, a reduced
food supply for pelagic larvae. However, this effect is
likely compounded by the fact that the time of greatest
primary productivity off the southwestern Australian
coast (autumn/winter) (Lourey et al. 2006) occurs at a
time of potentially unfavourable southward transport
for shelf spawning species. Clupeiod larvae are generally poor swimmers compared to other teleost larvae
(Fisher et al. 2005); their reproductive strategy instead
focuses on spawning throughout the year, so that
larvae hatched during times of favourable retention
may survive (Hutchings et al. 2002).
Current velocities within the Leeuwin Current
during autumn and winter regularly exceed 1 knot
(Smith et al. 1991), producing a potential southward
transport of larvae within the current of more than
40 km d–1 (assuming passive transport), or 20 km d–1
inshore of the main current as it extends inshore over
the shelf (Caputi et al. 1996), with no apparent mechanism for larval retention close to spawning areas on the
open continental shelf. The existence of apparently
separate sardine stocks between the west and south
coasts of Western Australia (Gaughan et al. 2001c) suggests that larvae entrained within the Leeuwin Current
are not usually recruited to south coast stocks. However, young (< 2 yr) sardines are reportedly common in
the comparatively protected waters of Geographe Bay,
north of Cape Naturaliste (32° 31’ S to 33° 32’ S, see
Fig 1) (Gaughan et al. 2001c). Dispersal of larvae
spawned on the open shelf around Two Rocks and
Perth to Geographe Bay is therefore possible.
165
Bakun (1996) proposed 3 major classes of physical
processes that characterise favourable reproductive
habitats for coastal pelagic fishes: enrichment processes (e.g. upwelling), concentration processes (e.g.
fronts and stable water columns), and retention mechanisms. All 3 mechanisms are largely absent from
southwestern Australian waters. There are no largescale upwellings, few concentration processes, usually
no strong hydrographic fronts, and no retention mechanisms for pelagic, shelf-spawned larvae during times
of higher productivity (autumn and winter). A combination of these factors may be responsible for the low
clupeiform stock sizes and catches off southwestern
Australia.
The hypothesis that small sizes of clupeiod stocks
and fisheries off southwestern Australia are largely
related to advective processes, rather than to food
limitation alone, is supported by our determination of
larval sardine growth rates. Although larval sardine
growth rates are highly variable worldwide, growth
rates of sardine larvae from the Two Rocks transect
were comparable to, or higher than, growth of similarly
aged larvae from other more productive parts of the
world (Castillo et al. 1985, Butler 1987). The surprisingly high growth rates of larval sardines in the current
study may have resulted from warm water temperatures provided by the Leeuwin Current. After food
availability, water temperature is the major determining factor for larval growth rate, as larvae typically
grow faster with increasing temperature up to an optimum level, above which growth declines (Oozeki &
Watanabe 2000). Few published data exist on seasonal
patterns of secondary production and biomass in the
region studied; however, existing data suggest that
maxima, if present, are in autumn/winter, similar to
patterns of chlorophyll biomass (Koslow et al. 2006).
There is thus no reason at present to suspect that high
larval growth rates are supported by secondary production peaks at other times of year.
The effect of deleterious advection processes on sardine populations has also been noted on the Agulhas
Bank, and in the Benguela Current system off southern
Africa, where, despite high primary productivity,
yields of pelagic fishes are lower than in the Humboldt
system off South America. This may be due in part to
the poor retention conditions for pelagic larvae along
the southern African coast, as a result of the strong
western boundary current (Agulhas Current) and the
strong upwelling environment present off the west
coast (Benguela system), which together may result in
considerable offshore losses of pelagic eggs and larvae
(Hutchings et al. 1998, 2002).
Overall, sardine eggs and larvae were found across
the continental shelf between Two Rocks and Cape
Naturaliste, and monthly transect data suggest spawn-
Mar Ecol Prog Ser 364: 157–167, 2008
166
ing activity throughout the year. However, while GSI
data suggest a winter spawning peak, egg and larval
concentrations were lower than expected in winter and
were higher in summer. As most sardine eggs and
larvae were sampled within the southward flowing
Leeuwin Current, which is strongest during winter,
this discrepancy may result from the significant southward advection of eggs and larvae during this time.
The low stock sizes and fisheries for sardine and other
clupeiod species off southwestern Australia may be
due to a combination of low primary productivity
(because of the suppression of large-scale upwelling
by the Leeuwin Current) compounded by a modest
seasonal maximum in primary productivity during the
time of least favourable retention for pelagic larvae
(i.e. winter). The relatively high growth rates of sardine larvae from the study area (by world standards)
support the notion that clupeiod fish populations are
controlled by a combination of advective processes in
conjunction with primary productivity patterns, rather
than simply by a scarcity of food for pelagic larvae.
Acknowledgements. The Western Australian Strategic
Research Fund for the Marine Environment and Murdoch
University funded part of this research and provided a PhD
scholarship for B.A.M. N. Mortimer, J. Strzelecki and T.
Koslow are thanked for coordinating sample collection on the
Two Rocks transect and for help with data interpretation. The
Western Australian Department of Fisheries is acknowledged
for sardine GSI, egg and larval sample collection and data
processing. The National Tidal Centre, the Australian Bureau
of Meteorology and DISC(NASA) provided oceanographic
and remotely sensed data, and A. Pearce and P. Fearns
helped with processing and interpretation of these data. M.
Harvey is thanked for help with GIS analyses.
➤
➤
➤
➤
➤
➤
➤
➤
➤
LITERATURE CITED
➤
➤
Bailey KM, Houde ED (1989) Predation on eggs and larvae of
marine fishes and the recruitment problem. Adv Mar Biol
25:1–83
Bakun A (1996) Patterns in the ocean: ocean processes and
marine population dynamics. California Sea Grant College System, University of California, La Jolla, CA
Beckley LE, Hewitson JD (1994) Distribution and abundance
of clupeiod larvae along the east coast of South Africa in
1990/91. S Afr J Mar Sci 14:205–212
Beckley LE, van der Lingen C (1999) Biology, fishery and
management of sardines (Sardinops sagax) in southern
African waters. Mar Freshw Res 50:955–978
Butler JL (1987) Comparison of the early life history parameters of Pacific sardine and northern anchovy with implications of species interactions. PhD thesis, University of
California, San Diego, CA
Caputi N, Fletcher WJ, Pearce AF, Chubb CF (1996) Effect of
the Leeuwin Current on the recruitment of fish and invertebrates along the Western Australian coast. Mar Freshw
Res 47:147–155
Castillo G, Aguilera G, Herrera G, Bernal PA and others
(1985) Larval growth rates of Pacific sardine Sardinops
➤
➤
➤
sagax off central Chile, determined by count of daily rings
in otoliths. Biol Pesq 14:3–10
Cerrato RM (1990) Interpretable statistical tests for growth
comparisons using parameters in the von Bertalanffy
equation. Can J Fish Aquat Sci 47:1416–1426
Cole J (1999) Environmental conditions, satellite imagery,
and clupeoid recruitment in the northern Benguela
upwelling system. Fish Oceanogr 8:25–38
DeVlamming V, Grossman G, Chapman F (1982) On the use
of the gonadosomatic index. Comp Biochem Physiol 73:
31–39
Fearns PR, Twomey L, Zakiyah U, Helleren S, Vincent W,
Lynch MJ (2007) The Hillarys Transect (3): optical and
chlorophyll relationships across the continental shelf off
Perth. Cont Shelf Res 27:1719–1746
Feng M, Meyers G, Pearce A, Wijffels S (2003) Annual and
interannual variations of the Leeuwin Current at 32°S.
J Geophys Res 108, doi:10.1029/2002JC001763
Fisher R, Leis JM, Clark DL, Wilson SK (2005) Critical swimming speeds of late-stage coral reef fish larvae: variation
within species, amongst species and between locations.
Mar Biol 147:1201–1212
Fletcher WJ (1999) Vertical distribution of pilchard (Sardinops
sagax) eggs and larvae off Southern Australia. Mar
Freshw Res 50:117–122
Fletcher WJ, Tregonning RJ (1992) Distribution and timing of
spawning by the Australian pilchard (Sardinops sagax
neopilchardus) off Albany, Western Australia. Aust J Mar
Freshw Res 43:1437–1449
Fletcher WJ, Tregonning RJ, Sant GJ (1994) Interseasonal
variation in the transport of pilchard eggs and larvae off
southern Western Australia. Mar Ecol Prog Ser 111:
209–224
Funes-Rodriguez R, Hinojosa-Medina A, Avendano-Ibarra R,
Hernandez-Rivas M, Saldierna-Martinez R, Watson W
(2001) Spawning of small pelagic fishes in Bahia Magdalena Baja California Sur, Mexico, at the beginning of
the 1997–1998 El Niño. Estuar Coast Shelf Sci 53:653–664
Gaughan DJ, Neira FJ, Beckley LE, Potter IC (1990) Composition, seasonality and distribution of the ichthyoplankton in
the lower Swan Estuary, south-western Australia. Aust J
Mar Freshw Res 41:529–543
Gaughan DJ, White KV, Fletcher WJ (2001a) The links
between functionally distinct adult assemblages of Sardinops sagax : larval advection across management
boundaries. ICES J Mar Sci 58:597–606
Gaughan DJ, Fletcher WJ, White KV (2001b) Growth rate of
larval Sardinops sagax from ecosystems with different
levels of productivity. Mar Biol 139:831–837
Gaughan DJ, Baudains GA, Mitchell RWD, Leary TI (2001c)
Pilchard (Sardinops sagax) nursery areas and recruitment
process assessment between different regions in southern
Western Australia. Fisheries, Research Report 131,
Department of Fisheries, Perth, Western Australia
Gersbach GH, Pattiaratchi C, Ivey GN, Cresswell GR (1999)
Upwelling on the south-west coast of Australia — source of
the Capes Current? Cont Shelf Res 19:363–400
Hanson CE, Pattiaratchi C, Waite AM (2005) Seasonal production regimes of south-western Australia: influence of
the Capes and Leeuwin Currents on phytoplankton
dynamics. Mar Freshw Res 56:1011–1026
Hayashi A, Yamashita Y, Kawaguchi K, Ishii T (1989) Rearing
method and daily otolith ring of Japanese sardine larvae.
Nippon Suisan Gakkai 55:997–1000
Hilborn R, Mangel M (1997) The ecological detective: confronting models with data. Princeton University Press,
Princeton, NJ
Muhling et al.: Sardine spawning, larval abundance and growth
➤
➤
➤
➤
➤
Huggett JA, Boyd AJ, Hutchings L, Kemp A (1998) Weekly
variability of clupeoid eggs and larvae in the Benguela jet
current: implications for recruitment. S Afr J Mar Sci
19:197–210
Hutchings L, Barange M, Bloomer SF, Boyd AJ and others
(1998) Multiple factors affecting South African anchovy
recruitment in the spawning transport and nursery areas.
S Afr J Mar Sci 19:211–225
Hutchings L, Beckley LE, Griffiths MH, Roberts MJ, Sundby
S, van der Lingen C (2002) Spawning on the edge: spawning grounds and nursery areas around the southern
African coastline. Mar Freshw Res 53:307–318
Hutchins B, Thompson M (2001) Marine and estuarine fishes
of south-western Australia: a field guide for anglers and
divers. Western Australian Museum, Perth
Koslow JA, Pearce AF, Mortimer N, Strzelecki J and others
(2006) Biophysical oceanography off Western Australia: biological communities, pattern, process, and ecosystem dynamics across the continental shelf and slope. In: Keesing
JK, Heine JN (eds) Strategic Research Fund for the Marine
Environment Final Report, Vol 1. The SRFME initiative and
collaborative linkages program. Strategic Research Fund for
the Marine Environment, CSIRO, Perth, p 143–213
Lasker R, Smith PE (1976) Estimation of the effects of environmental variations on the eggs and larvae of the northern
anchovy. CCOFI Rep 19:128–137
Lourey MJ, Dunn JR, Waring J (2006) A mixed-layer climatology of Leeuwin Current and Western Australian shelf
waters: seasonal nutrient dynamics and biomass. J Mar
Syst 59:25–51
Muhling BA, Beckley LE (2007) Seasonal variation in horizontal and vertical structure of larval fish assemblages off
south-western Australia, with implications for larval
transport. J Plankton Res 29:967–983
Muhling BA, Beckley LE, Koslow JA, Pearce AF (2008) Larval
fish assemblages and water mass structure off the oligotrophic south-western Australian coast. Fish Oceanogr
17:16–31
Olivar MP, Salat J, Palomera I (2001) Comparative study of
spatial distribution patterns of early stages of anchovy and
pilchard in the NW Mediterranean Sea. Mar Ecol Prog Ser
217:111–120
Oozeki Y, Watanabe Y (2000) Comparison of somatic growth
and otolith increment growth in laboratory-reared larvae
of Pacific saury, Cololabis saira, under different temperaEditorial responsibility: Matthias Seaman,
Oldendorf/Luhe, Germany
167
ture conditions. Mar Biol 136:349–359
➤ Pearce AF, Pattiaratchi C (1999) The Capes Current: a sum➤
➤
➤
➤
➤
➤
➤
mer countercurrent flowing past Cape Leeuwin and Cape
Naturaliste, Western Australia. Cont Shelf Res 19:401–420
Pearce AF, Lynch MJ, Hanson CE (2006) The Hillarys transect
(1): seasonal and cross-shelf variability of physical and
chemical water properties off Perth, Western Australia,
1996–98. Cont Shelf Res 26:1689–1729
Sanchez-Velaso L, Avalos-Garcia C, Renteria-Cano M, Shirasago B (2004) Fish larvae abundance and distribution in
the central Gulf of California during strong environmental
changes (1997–1998 El Niño and 1998–1999 La Niña).
Deep-Sea Res 51:711–722
Schwartzlose RA, Alheit J, Bakun A, Baumgartner TR and
others (1999) Worldwide large-scale fluctuations of sardine and anchovy populations. S Afr J Mar Sci 21:289–348
Smith PE, Moser HG (2003) Long-term trends and variability
in the larvae of Pacific sardine and associated fish species
of the California Current Region. Deep-Sea Res 50:
2519–2536
Smith RL, Huyer A, Godfrey JS, Church JA (1991) The
Leeuwin Current off Western Australia, 1986–87. J Phys
Oceanogr 21:323–345
Somarakis S, Ganias K, Siapatis A, Koutsikopoulos C,
Machias A, Papaconstantinou C (2006) Spawning habitat
and daily egg production of sardine (Sardina pilchardus)
in the eastern Mediterranean. Fish Oceanogr 15:281–292
Stevenik EK, Sundby S, Cloete R (2001) Influence of buoyancy and vertical distribution of sardine Sardinops sagax
eggs and larvae on their transport in the northern
Benguela ecosystem. S Afr J Mar Sci 23:85–98
Thomas AC, Strub TP, Carr ME, Weatherbee R (2004) Comparisons of chlorophyll variability between the four major
global eastern boundary currents. Int J Remote Sens 25:
1443–1447
Ward TM, Staunton-Smith J, Hoyle S, Halliday IA (2003)
Spawning patterns of four species of predominantly temperate pelagic fishes in the sub-tropical waters of southern
Queensland. Estuar Coast Shelf Sci 56:1125–1140
Ward TM, McLeay LJ, Dimmlich WF, Rogers PJ and others
(2006) Pelagic ecology of a northern boundary current system: effects of upwelling on the production and distribution of sardine (Sardinops sagax), anchovy (Engraulis australis) and southern bluefin tuna (Thunnus maccoyii ) in
the Great Australian Bight. Fish Oceanogr 15:191–207
Submitted: January 4, 2007; Accepted: March 13, 2008
Proofs received from author(s): July 3, 2008